Causal information alone does not enable children to revise their beliefs about improbable events

• In Event: Belief revision in a social world: How other people shape children’s ability to revise beliefs

Abstract

Adults typically believe that strange and improbable events, like eating pickle-flavored ice-cream, are almost as possible as everyday events like wearing a hat or washing a car. Young children, however, often deny that improbable events are possible (Shtulman & Carey, 2007; Shutlman, 2009). Recent work also suggests that these beliefs are difficult to revise. For instance, even when explicitly told that improbable events can happen, children persist in denying this (Lane et al., 2018).

In two experiments, we explore whether belief revision about improbable events might be enabled if children are provided with causal information about how the events could happen. This proposal is inspired by claims that children deny events are possible because they fail to imagine circumstances that would allow the events to occur (Lane et al., 2016; Shtulman, 2009). We also explore whether belief revision might be prompted by informing children about the occurrence of similar events. This would suggest that children instead reason about possibility by retrieving instances of similar events from memory (Goulding & Friedman, 2020). These accounts are pitted against each other to reveal their relative influence on children’s beliefs about possibility.

In Experiment 1, 4-7-year-olds (N=240, Mage=5;11) heard either causal or noncausal information about three ordinary events. For example, children learned a process for making strawberry ice-cream (causal) or learned about places where they could get strawberry ice-cream (noncausal). Importantly, the causal information provided was broad enough that it could potentially enable a variety of outcomes. Children then judged whether a related improbable event, ordinary event, and impossible event could happen. Children differentiated between all three event-types, ps<.001, but their judgments were not influenced by the kind of information provided, p=.787 (see Figure 1). This suggests that causal information had little impact on children’s beliefs about possibility.

In Experiment 2, 5-6-year-olds (N=109, Mage=6;0) heard one of three different kinds of information: causal information about an ordinary event, causal information about a similar improbable event, or the mere fact that a similar improbable event happened. Crucially, the causal information was identical for both kinds of events. Children then judged whether a related improbable event and impossible event could happen. Children more often agreed that improbable and impossible events could happen after hearing causal information about similar improbable events than if they heard identical causal information about ordinary events, p<.001. Also, children agreed significantly more after hearing causal information about improbable events than after merely hearing that the improbable events happened, p=.044 (see Figure 2).

Our findings suggest that causal information often plays a weak role in promoting children’s belief revision about the possibility of improbable events, and that they may instead be more influenced by the knowledge that a similar improbable event occurred. However, we also found that children were most likely to affirm improbable events after hearing causal knowledge about a similar event. This suggests that causal knowledge may increase children’s willingness to accept the possibility of an improbable event after a similar event has already been retrieved from memory.

Authors

• Brandon Wade Goulding, University of Winnipeg
  Presenting Author

• Ori Friedman, University of Waterloo
  Non-Presenting Author